Abstract
This study explores the possibility of using a machine learning approach to analysing social media big data for tourism demand forecasting. We demonstrate how to extract the main topics discussed on Twitter and calculate the mean sentiment score for each topic as the proxy of the general attitudes towards those topics, which are then used for predicting tourist arrivals. We choose Sydney, Australia as the case for testing the performance and validity of our proposed forecasting framework. The study reveals key topics discussed in social media that can be used to predict tourist arrivals in Sydney. The study has both theoretical implications for tourist behavioural research and practical implications for destination marketing.
Original language | English |
---|---|
Pages (from-to) | 32-43 |
Number of pages | 12 |
Journal | Journal of Digital Economy |
Volume | 1 |
Issue number | 1 |
Early online date | 27 Aug 2022 |
DOIs | |
Publication status | Published - 2022 |
Externally published | Yes |
Keywords
- Natural language processing
- Machine learning
- Social listening
- Tourist attitude
- Tourist arrival
- Tourism demand forecasting